/* Adjust space between sidebar sections */
.sidebar > .section {
    margin-bottom: 0px; /* Adjust spacing as needed */
  }
  
  /* Adjust space between pages within a section */
  .sidebar .page {
    margin-bottom: 0px; /* Adjust spacing as needed */
  }

  /* Adjust padding inside each page link */
.sidebar .page a {
    padding: 0px 0px; /* Adjust top/bottom and left/right padding */
  }
  
  /* Section highlight animation when targeted through anchor links */
  :target,
  h2[id]:target,
  h3[id]:target,
  h4[id]:target,
  h5[id]:target,
  div[id]:target {
    position: relative;
    background: linear-gradient(to left, #e9b405, #2b2a2a00);
    animation: highlight-section 3s ease-in-out;
    scroll-margin-top: 80px;
    padding: 0.2rem;
    border-radius: 8px;
    border-left: 4px solid #f2ca46;
  }
  
  @keyframes highlight-section {
    0% {
      background: linear-gradient(to left, #e9b405, #2b2a2a00);
    }
    50% {
      background: linear-gradient(to left, #e9b405, #2b2a2a00);
    }
    100% {
      background: linear-gradient(to left, #e9b405, #2b2a2a00);
    }
  }

Cancel an evaluation. Can cancel queued and in_progress evaluations

Cancel evaluation

BearerAuth

DeepRails

DeepRails Overview

Learn why LLM-as-a-Judge is the new gold standard for evaluating AI responses

LLM Evaluations

Experiment and iterate with fast, automated AI evaluations across key quality dimensions

Evaluate Overview

Start using DeepRails Evaluate in a few minutes with this quickstart guide, and integrate it seamlessly into your existing workflow.

Quickstart Guide

Monitor and analyze AI application performance in production with real-time insights to maintain quality outputs

Monitor Overview

Start using DeepRails Monitor to observe how users interact with your generative AI application in production, and surface quality issues before they escalate.

Safeguard your AI applications in production using DeepRails Defend—powered by real-time guardrails and intelligent quality filters.

Defend Overview

The next evolution in LLM evaluation - multi-model collective intelligence for superior accuracy

HyperChainpoll

How DeepRails aggregates granular evaluations into actionable scores

Scoring Methodology

Choose your evaluation accuracy vs. cost tradeoff with Run Modes

Run Modes

Explore DeepRails Guardrail metrics designed to evaluate LLM behavior.

Metrics Comparison

Evaluate factual accuracy in AI outputs using DeepRails Guardrail Metrics to detect and prevent hallucinations in your AI systems.

Correctness

Evaluate whether AI responses thoroughly and accurately address all aspects of a user's query using DeepRails Guardrail Metrics.

Completeness

Evaluate the safety of AI-generated content using DeepRails Guardrail Metrics to identify and mitigate harmful or high-risk responses.

API Documentation

Cancel evaluation

Authorizations

Path Parameters

Response